
import pandas as pd
import numpy as np
import seaborn as sns
from sklearn.tree import DecisionTreeClassifier
from sklearn import metrics
from sklearn import tree
from sklearn.ensemble import RandomForestClassifier


df_bcancer=pd.read_csv("d:/csv-ml/breast_cancer.csv")


df_bcancer.head()


#number of missing values


df_bcancer.isna().sum()

id                         0
diagnosis                  0
radius_mean                0
texture_mean               0
perimeter_mean             0
area_mean                  0
smoothness_mean            0
compactness_mean           0
concavity_mean             0
concave points_mean        0
symmetry_mean              0
fractal_dimension_mean     0
radius_se                  0
texture_se                 0
perimeter_se               0
area_se                    0
smoothness_se              0
compactness_se             0
concavity_se               0
concave points_se          0
symmetry_se                0
fractal_dimension_se       0
radius_worst               0
texture_worst              0
perimeter_worst            0
area_worst                 0
smoothness_worst           0
compactness_worst          0
concavity_worst            0
concave points_worst       0
symmetry_worst             0
fractal_dimension_worst    0
dtype: int64


df_bcancer.dtypes

id                           int64
diagnosis                   object
radius_mean                float64
texture_mean               float64
perimeter_mean             float64
area_mean                  float64
smoothness_mean            float64
compactness_mean           float64
concavity_mean             float64
concave points_mean        float64
symmetry_mean              float64
fractal_dimension_mean     float64
radius_se                  float64
texture_se                 float64
perimeter_se               float64
area_se                    float64
smoothness_se              float64
compactness_se             float64
concavity_se               float64
concave points_se          float64
symmetry_se                float64
fractal_dimension_se       float64
radius_worst               float64
texture_worst              float64
perimeter_worst            float64
area_worst                 float64
smoothness_worst           float64
compactness_worst          float64
concavity_worst            float64
concave points_worst       float64
symmetry_worst             float64
fractal_dimension_worst    float64
dtype: object


x=df_bcancer.drop(["diagnosis","id"],axis=1)


x.head()


x.columns

Index(['radius_mean', 'texture_mean', 'perimeter_mean', 'area_mean',
       'smoothness_mean', 'compactness_mean', 'concavity_mean',
       'concave points_mean', 'symmetry_mean', 'fractal_dimension_mean',
       'radius_se', 'texture_se', 'perimeter_se', 'area_se', 'smoothness_se',
       'compactness_se', 'concavity_se', 'concave points_se', 'symmetry_se',
       'fractal_dimension_se', 'radius_worst', 'texture_worst',
       'perimeter_worst', 'area_worst', 'smoothness_worst',
       'compactness_worst', 'concavity_worst', 'concave points_worst',
       'symmetry_worst', 'fractal_dimension_worst'],
      dtype='object')


#analysing the features using pair plot


sns.pairplot(df_bcancer,hue="diagnosis")

<seaborn.axisgrid.PairGrid at 0x118b8eda4c0>


y=df_bcancer.loc[:,"diagnosis"]


#printing the correlation of the features


df_cor=x.corr()


sns.heatmap(df_cor)

<AxesSubplot:>


#filtering the rows which have high multicolinearity


df_cor[(df_cor>=.6)| (df_cor<-6.0)]


#calculate multicollinearity using VIF


from statsmodels.stats.outliers_influence import variance_inflation_factor


df_vif=pd.DataFrame()


df_vif["features"]=x.columns


df_vif["vif"]=[variance_inflation_factor(x.values,i) for i in range(x.shape[1])]


df_vif


df_vif=df_vif.sort_values(["vif"],ascending=False)


df_vif


#selecting 2 best features by chi square


from sklearn.feature_selection import SelectKBest, chi2


fs_model=SelectKBest(chi2,k=2).fit(x,y)


cols=fs_model.get_support(indices=True)


cols

array([ 3, 23], dtype=int64)


x.head()


x_new=x.iloc[:,cols]


x_new.head()


from sklearn.neighbors import KNeighborsClassifier


from sklearn.model_selection import train_test_split
from sklearn.model_selection import GridSearchCV


x_tr,x_test,y_tr,y_test=train_test_split(x_new,y,test_size=.2,random_state=100)


#finding the best value of k


params={"n_neighbors":range(1,51)}


gv_mod=GridSearchCV(KNeighborsClassifier(),params).fit(x_tr,y_tr)


x_tr.shape

(455, 2)


gv_mod.best_params_

{'n_neighbors': 5}


kmodel1=KNeighborsClassifier(n_neighbors=gv_mod.best_params_["n_neighbors"]).fit(x_tr,y_tr)


p1=kmodel1.predict(x_test)


from sklearn.metrics import accuracy_score


accuracy_score(y_test,p1)

0.9473684210526315


df_bcancer.shape

(569, 32)


x.shape

(569, 30)


# select features using ANOVA


from sklearn.feature_selection import SelectKBest,f_classif


fs_mod2=SelectKBest(f_classif,k=2).fit(x,y)


cols=fs_mod2.get_support(indices=True)


cols

array([22, 27], dtype=int64)


x_new2=x.iloc[:,cols]


# create KNN model


x_tr,x_test,y_tr,y_test=train_test_split(x_new2,y,test_size=.2,random_state=100)


gs_mod2=GridSearchCV(KNeighborsClassifier(),params).fit(x_tr,y_tr)


gs_mod2.best_params_

{'n_neighbors': 22}


knn_mod2=KNeighborsClassifier(gs_mod2.best_params_["n_neighbors"]).fit(x_tr,y_tr)


p2=knn_mod2.predict(x_test)


accuracy_score(y_test,p2)

0.9385964912280702


from sklearn.decomposition import PCA


# scaling or normalisation


from sklearn.preprocessing import MinMaxScaler #x-min(x)/(max(x)-min(x))


scaled_x=MinMaxScaler().fit_transform(x)


scaled_x

array([[0.52103744, 0.0226581 , 0.54598853, ..., 0.91202749, 0.59846245,
        0.41886396],
       [0.64314449, 0.27257355, 0.61578329, ..., 0.63917526, 0.23358959,
        0.22287813],
       [0.60149557, 0.3902604 , 0.59574321, ..., 0.83505155, 0.40370589,
        0.21343303],
       ...,
       [0.45525108, 0.62123774, 0.44578813, ..., 0.48728522, 0.12872068,
        0.1519087 ],
       [0.64456434, 0.66351031, 0.66553797, ..., 0.91065292, 0.49714173,
        0.45231536],
       [0.03686876, 0.50152181, 0.02853984, ..., 0.        , 0.25744136,
        0.10068215]])


pca_x=PCA(n_components=.95).fit_transform(scaled_x,y)


pca_x.shape

(569, 10)


pca_mod=PCA(n_components=.95).fit(scaled_x,y)


pca_mod.explained_variance_

array([0.33133389, 0.10785038, 0.0443947 , 0.04000678, 0.02549742,
       0.01916637, 0.00986455, 0.00743488, 0.00616788, 0.00589966])


pca_list=pca_mod.explained_variance_


import matplotlib.pyplot as plt
plt.xlabel("number of components")
plt.ylabel("cumulative explained variance")
plt.plot(range(1,len(pca_list)+1),np.cumsum(pca_list),c="green",marker="o",markerfacecolor="red")

[<matplotlib.lines.Line2D at 0x118f1747940>]


pca_x=pca_mod.fit_transform(scaled_x)


x_tr,x_test,y_tr,y_test=train_test_split(pca_x,y,test_size=.2,random_state=100)


# best value of k


gs_mod2=GridSearchCV(KNeighborsClassifier(),params).fit(x_tr,y_tr)


gs_mod2.best_params_

{'n_neighbors': 7}


gs_mod2.best_params_["n_neighbors"]

7


k_mod3=KNeighborsClassifier(gs_mod2.best_params_["n_neighbors"]).fit(x_tr,y_tr)


p_knn_pca_tr=k_mod3.predict(x_tr)


metrics.recall_score(y_tr,p_knn_pca_tr,average="weighted")

0.9802197802197802


p3=k_mod3.predict(x_test)


accuracy_score(y_test,p3)

0.9649122807017544


from sklearn.metrics import confusion_matrix


confusion_matrix(y_test,p3)

array([[65,  0],
       [ 4, 45]], dtype=int64)


metrics.recall_score(y_test,p3,average="weighted")

0.9649122807017544


x_tr,x_test,y_tr,y_test=train_test_split(x,y,test_size=.2,random_state=100)


dtree=DecisionTreeClassifier(criterion="gini").fit(x_tr,y_tr)


p_tr_dtree=dtree.predict(x_tr)


metrics.confusion_matrix(y_tr,p_tr_dtree)

array([[292,   0],
       [  0, 163]], dtype=int64)


metrics.accuracy_score(y_tr,p_tr_dtree)

1.0


metrics.recall_score(y_tr,p_tr_dtree,average="weighted")

1.0


plt.figure(figsize=(50,50))
tree.plot_tree(dtree,filled=True,feature_names = x_tr.columns,class_names=True,fontsize=30)
plt.show()


p_test_dtree=dtree.predict(x_test)


metrics.recall_score(y_test,p_test_dtree,average="weighted")

0.956140350877193


x_tr,x_test,y_tr,y_test=train_test_split(x,y,test_size=.2,random_state=100)


rf_dtree=RandomForestClassifier(n_estimators=500).fit(x_tr,y_tr)


p_rf_tr=rf_dtree.predict(x_tr)


metrics.recall_score(y_tr,p_rf_tr,average="weighted")

1.0


metrics.accuracy_score(y_tr,p_rf_tr)

1.0


p_rf_test=rf_dtree.predict(x_test)


metrics.recall_score(y_test,p_rf_test,average="weighted")

0.9649122807017544


#The random forest model though shows higher performance shows overfitting (96.4%)


x_tr,x_test,y_tr,y_test=train_test_split(pca_x,y,test_size=.2,random_state=100)


rf_pca=RandomForestClassifier(n_estimators=500).fit(x_tr,y_tr)


pca_rf_tr=rf_pca.predict(x_tr)


metrics.recall_score(y_tr,pca_rf_tr,average="weighted")

1.0


metrics.accuracy_score(y_tr,pca_rf_tr)

1.0


pca_rf_test=rf_pca.predict(x_test)


metrics.recall_score(y_test,pca_rf_test,average="weighted")

0.9736842105263158


#The random forest model though shows higher performance shows overfitting (97.3%)


import xgboost as xgb


y.value_counts()

B    357
M    212
Name: diagnosis, dtype: int64


y_new=y.map({"B":0,"M":1})


y_new.value_counts()

0    357
1    212
Name: diagnosis, dtype: int64


xgb_mod=xgb.XGBClassifier(objective="reg:logistic",learning_rate=.1,n_estimators=100,max_depth=5,colsample_bytree=.3)


x_tr,x_test,y_tr,y_test=train_test_split(x,y_new,test_size=.2,random_state=100)


xgb_mod.fit(x_tr,y_tr)

XGBClassifier(base_score=0.5, booster='gbtree', callbacks=None,
              colsample_bylevel=1, colsample_bynode=1, colsample_bytree=0.3,
              early_stopping_rounds=None, enable_categorical=False,
              eval_metric=None, gamma=0, gpu_id=-1, grow_policy='depthwise',
              importance_type=None, interaction_constraints='',
              learning_rate=0.1, max_bin=256, max_cat_to_onehot=4,
              max_delta_step=0, max_depth=5, max_leaves=0, min_child_weight=1,
              missing=nan, monotone_constraints='()', n_estimators=100,
              n_jobs=0, num_parallel_tree=1, objective='reg:logistic',
              predictor='auto', random_state=0, reg_alpha=0, ...)

XGBClassifier(base_score=0.5, booster='gbtree', callbacks=None,
              colsample_bylevel=1, colsample_bynode=1, colsample_bytree=0.3,
              early_stopping_rounds=None, enable_categorical=False,
              eval_metric=None, gamma=0, gpu_id=-1, grow_policy='depthwise',
              importance_type=None, interaction_constraints='',
              learning_rate=0.1, max_bin=256, max_cat_to_onehot=4,
              max_delta_step=0, max_depth=5, max_leaves=0, min_child_weight=1,
              missing=nan, monotone_constraints='()', n_estimators=100,
              n_jobs=0, num_parallel_tree=1, objective='reg:logistic',
              predictor='auto', random_state=0, reg_alpha=0, ...)


p_xgb_tr=xgb_mod.predict(x_tr)


metrics.recall_score(y_tr,p_xgb_tr)

1.0


metrics.accuracy_score(y_tr,p_xgb_tr)

1.0


p_xgb_test=xgb_mod.predict(x_test)


metrics.recall_score(y_test,p_xgb_test)

0.9183673469387755


metrics.accuracy_score(y_test,p_xgb_test)

0.956140350877193

	id	diagnosis	radius_mean	texture_mean	perimeter_mean	area_mean	smoothness_mean	compactness_mean	concavity_mean	concave points_mean	...	radius_worst	texture_worst	perimeter_worst	area_worst	smoothness_worst	compactness_worst	concavity_worst	concave points_worst	symmetry_worst	fractal_dimension_worst
0	842302	M	17.99	10.38	122.80	1001.0	0.11840	0.27760	0.3001	0.14710	...	25.38	17.33	184.60	2019.0	0.1622	0.6656	0.7119	0.2654	0.4601	0.11890
1	842517	M	20.57	17.77	132.90	1326.0	0.08474	0.07864	0.0869	0.07017	...	24.99	23.41	158.80	1956.0	0.1238	0.1866	0.2416	0.1860	0.2750	0.08902
2	84300903	M	19.69	21.25	130.00	1203.0	0.10960	0.15990	0.1974	0.12790	...	23.57	25.53	152.50	1709.0	0.1444	0.4245	0.4504	0.2430	0.3613	0.08758
3	84348301	M	11.42	20.38	77.58	386.1	0.14250	0.28390	0.2414	0.10520	...	14.91	26.50	98.87	567.7	0.2098	0.8663	0.6869	0.2575	0.6638	0.17300
4	84358402	M	20.29	14.34	135.10	1297.0	0.10030	0.13280	0.1980	0.10430	...	22.54	16.67	152.20	1575.0	0.1374	0.2050	0.4000	0.1625	0.2364	0.07678

	radius_mean	texture_mean	perimeter_mean	area_mean	smoothness_mean	compactness_mean	concavity_mean	concave points_mean	symmetry_mean	fractal_dimension_mean	...	radius_worst	texture_worst	perimeter_worst	area_worst	smoothness_worst	compactness_worst	concavity_worst	concave points_worst	symmetry_worst	fractal_dimension_worst
0	17.99	10.38	122.80	1001.0	0.11840	0.27760	0.3001	0.14710	0.2419	0.07871	...	25.38	17.33	184.60	2019.0	0.1622	0.6656	0.7119	0.2654	0.4601	0.11890
1	20.57	17.77	132.90	1326.0	0.08474	0.07864	0.0869	0.07017	0.1812	0.05667	...	24.99	23.41	158.80	1956.0	0.1238	0.1866	0.2416	0.1860	0.2750	0.08902
2	19.69	21.25	130.00	1203.0	0.10960	0.15990	0.1974	0.12790	0.2069	0.05999	...	23.57	25.53	152.50	1709.0	0.1444	0.4245	0.4504	0.2430	0.3613	0.08758
3	11.42	20.38	77.58	386.1	0.14250	0.28390	0.2414	0.10520	0.2597	0.09744	...	14.91	26.50	98.87	567.7	0.2098	0.8663	0.6869	0.2575	0.6638	0.17300
4	20.29	14.34	135.10	1297.0	0.10030	0.13280	0.1980	0.10430	0.1809	0.05883	...	22.54	16.67	152.20	1575.0	0.1374	0.2050	0.4000	0.1625	0.2364	0.07678

	radius_mean	texture_mean	perimeter_mean	area_mean	smoothness_mean	compactness_mean	concavity_mean	concave points_mean	symmetry_mean	fractal_dimension_mean	...	radius_worst	texture_worst	perimeter_worst	area_worst	smoothness_worst	compactness_worst	concavity_worst	concave points_worst	symmetry_worst	fractal_dimension_worst
radius_mean	1.000000	NaN	0.997855	0.987357	NaN	NaN	0.676764	0.822529	NaN	NaN	...	0.969539	NaN	0.965137	0.941082	NaN	NaN	NaN	0.744214	NaN	NaN
texture_mean	NaN	1.000000	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	...	NaN	0.912045	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
perimeter_mean	0.997855	NaN	1.000000	0.986507	NaN	NaN	0.716136	0.850977	NaN	NaN	...	0.969476	NaN	0.970387	0.941550	NaN	NaN	NaN	0.771241	NaN	NaN
area_mean	0.987357	NaN	0.986507	1.000000	NaN	NaN	0.685983	0.823269	NaN	NaN	...	0.962746	NaN	0.959120	0.959213	NaN	NaN	NaN	0.722017	NaN	NaN
smoothness_mean	NaN	NaN	NaN	NaN	1.000000	0.659123	NaN	NaN	NaN	NaN	...	NaN	NaN	NaN	NaN	0.805324	NaN	NaN	NaN	NaN	NaN
compactness_mean	NaN	NaN	NaN	NaN	0.659123	1.000000	0.883121	0.831135	0.602641	NaN	...	NaN	NaN	NaN	NaN	NaN	0.865809	0.816275	0.815573	NaN	0.687382
concavity_mean	0.676764	NaN	0.716136	0.685983	NaN	0.883121	1.000000	0.921391	NaN	NaN	...	0.688236	NaN	0.729565	0.675987	NaN	0.754968	0.884103	0.861323	NaN	NaN
concave points_mean	0.822529	NaN	0.850977	0.823269	NaN	0.831135	0.921391	1.000000	NaN	NaN	...	0.830318	NaN	0.855923	0.809630	NaN	0.667454	0.752399	0.910155	NaN	NaN
symmetry_mean	NaN	NaN	NaN	NaN	NaN	0.602641	NaN	NaN	1.000000	NaN	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0.699826	NaN
fractal_dimension_mean	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	1.000000	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0.767297
radius_se	0.679090	NaN	0.691765	0.732562	NaN	NaN	0.631925	0.698050	NaN	NaN	...	0.715065	NaN	0.719684	0.751548	NaN	NaN	NaN	NaN	NaN	NaN
texture_se	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
perimeter_se	0.674172	NaN	0.693135	0.726628	NaN	NaN	0.660391	0.710650	NaN	NaN	...	0.697201	NaN	0.721031	0.730713	NaN	NaN	NaN	NaN	NaN	NaN
area_se	0.735864	NaN	0.744983	0.800086	NaN	NaN	0.617427	0.690299	NaN	NaN	...	0.757373	NaN	0.761213	0.811408	NaN	NaN	NaN	NaN	NaN	NaN
smoothness_se	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
compactness_se	NaN	NaN	NaN	NaN	NaN	0.738722	0.670279	NaN	NaN	NaN	...	NaN	NaN	NaN	NaN	NaN	0.678780	0.639147	NaN	NaN	NaN
concavity_se	NaN	NaN	NaN	NaN	NaN	NaN	0.691270	NaN	NaN	NaN	...	NaN	NaN	NaN	NaN	NaN	NaN	0.662564	NaN	NaN	NaN
concave points_se	NaN	NaN	NaN	NaN	NaN	0.642262	0.683260	0.615634	NaN	NaN	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0.602450	NaN	NaN
symmetry_se	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
fractal_dimension_se	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0.688132	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
radius_worst	0.969539	NaN	0.969476	0.962746	NaN	NaN	0.688236	0.830318	NaN	NaN	...	1.000000	NaN	0.993708	0.984015	NaN	NaN	NaN	0.787424	NaN	NaN
texture_worst	NaN	0.912045	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	...	NaN	1.000000	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
perimeter_worst	0.965137	NaN	0.970387	0.959120	NaN	NaN	0.729565	0.855923	NaN	NaN	...	0.993708	NaN	1.000000	0.977578	NaN	NaN	0.618344	0.816322	NaN	NaN
area_worst	0.941082	NaN	0.941550	0.959213	NaN	NaN	0.675987	0.809630	NaN	NaN	...	0.984015	NaN	0.977578	1.000000	NaN	NaN	NaN	0.747419	NaN	NaN
smoothness_worst	NaN	NaN	NaN	NaN	0.805324	NaN	NaN	NaN	NaN	NaN	...	NaN	NaN	NaN	NaN	1.000000	NaN	NaN	NaN	NaN	0.617624
compactness_worst	NaN	NaN	NaN	NaN	NaN	0.865809	0.754968	0.667454	NaN	NaN	...	NaN	NaN	NaN	NaN	NaN	1.000000	0.892261	0.801080	0.614441	0.810455
concavity_worst	NaN	NaN	NaN	NaN	NaN	0.816275	0.884103	0.752399	NaN	NaN	...	NaN	NaN	0.618344	NaN	NaN	0.892261	1.000000	0.855434	NaN	0.686511
concave points_worst	0.744214	NaN	0.771241	0.722017	NaN	0.815573	0.861323	0.910155	NaN	NaN	...	0.787424	NaN	0.816322	0.747419	NaN	0.801080	0.855434	1.000000	NaN	NaN
symmetry_worst	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0.699826	NaN	...	NaN	NaN	NaN	NaN	NaN	0.614441	NaN	NaN	1.000000	NaN
fractal_dimension_worst	NaN	NaN	NaN	NaN	NaN	0.687382	NaN	NaN	NaN	0.767297	...	NaN	NaN	NaN	NaN	0.617624	0.810455	0.686511	NaN	NaN	1.000000

	radius_mean	texture_mean	perimeter_mean	area_mean	smoothness_mean	compactness_mean	concavity_mean	concave points_mean	symmetry_mean	fractal_dimension_mean	...	radius_worst	texture_worst	perimeter_worst	area_worst	smoothness_worst	compactness_worst	concavity_worst	concave points_worst	symmetry_worst	fractal_dimension_worst
0	17.99	10.38	122.80	1001.0	0.11840	0.27760	0.3001	0.14710	0.2419	0.07871	...	25.38	17.33	184.60	2019.0	0.1622	0.6656	0.7119	0.2654	0.4601	0.11890
1	20.57	17.77	132.90	1326.0	0.08474	0.07864	0.0869	0.07017	0.1812	0.05667	...	24.99	23.41	158.80	1956.0	0.1238	0.1866	0.2416	0.1860	0.2750	0.08902
2	19.69	21.25	130.00	1203.0	0.10960	0.15990	0.1974	0.12790	0.2069	0.05999	...	23.57	25.53	152.50	1709.0	0.1444	0.4245	0.4504	0.2430	0.3613	0.08758
3	11.42	20.38	77.58	386.1	0.14250	0.28390	0.2414	0.10520	0.2597	0.09744	...	14.91	26.50	98.87	567.7	0.2098	0.8663	0.6869	0.2575	0.6638	0.17300
4	20.29	14.34	135.10	1297.0	0.10030	0.13280	0.1980	0.10430	0.1809	0.05883	...	22.54	16.67	152.20	1575.0	0.1374	0.2050	0.4000	0.1625	0.2364	0.07678

	area_mean	area_worst
0	1001.0	2019.0
1	1326.0	1956.0
2	1203.0	1709.0
3	386.1	567.7
4	1297.0	1575.0

Classification using breast cancer dataset¶

many of the features have high multicollinearity indicate redundancy and may lead to overfitting,so we are elliminating certain features¶

Feature selection¶

The two best features selected¶

Create KNN model¶

Accuracy score using feature selection(chi2) and KNN is 96%¶

Accuracy score using feature selection(ANOVA) and KNN is 93%¶

PCA diensional reduction technique¶

PCA is giving the highest accuracy using KNN algorithm 96%¶

Decision tree model¶

Random forest model¶

Random forest model with PCA¶

XGBoost Classifier model¶

Conclusion¶

	features	vif
0	radius_mean	63306.172036
1	texture_mean	251.047108
2	perimeter_mean	58123.586079
3	area_mean	1287.262339
4	smoothness_mean	393.398166
5	compactness_mean	200.980354
6	concavity_mean	157.855046
7	concave points_mean	154.241268
8	symmetry_mean	184.426558
9	fractal_dimension_mean	629.679874
10	radius_se	236.665738
11	texture_se	24.675367
12	perimeter_se	211.396334
13	area_se	72.466468
14	smoothness_se	26.170243
15	compactness_se	44.919651
16	concavity_se	33.244099
17	concave points_se	53.698656
18	symmetry_se	37.176452
19	fractal_dimension_se	27.532631
20	radius_worst	9674.742602
21	texture_worst	343.004387
22	perimeter_worst	4487.781270
23	area_worst	1138.759252
24	smoothness_worst	375.597155
25	compactness_worst	132.884276
26	concavity_worst	86.310362
27	concave points_worst	148.673180
28	symmetry_worst	218.919805
29	fractal_dimension_worst	423.396723

Classification using breast cancer dataset¶

many of the features have high multicollinearity indicate redundancy and may lead to overfitting,so we are elliminating certain features¶

Feature selection¶

The two best features selected¶

Create KNN model¶

Accuracy score using feature selection(chi2) and KNN is 96%¶

Accuracy score using feature selection(ANOVA) and KNN is 93%¶

PCA diensional reduction technique¶

PCA is giving the highest accuracy using KNN algorithm 96%¶

Decision tree model¶

Random forest model¶

Random forest model with PCA¶

XGBoost Classifier model¶

Conclusion¶

After analysing all models, I recommend implementing KNN with PCA as it shows high performance without overfitting¶